Civil engineering structures hold relevant social, economic, and strategic importance. Many of them exceed their estimated design life and must withstand degradation, damage, and unforeseen natural events. In this context, implementing efficient Structural Health Monitoring (SHM) strategies ensures adequate performance levels, efficient maintenance interventions planning and extends structures' service life. This thesis focuses on the SHM of an existing industrial building subjected to the excitation produced by injection moulding machines. Their working cycles generate harmonic components at frequencies close to those characterising the building's dynamic behaviour. The analysis aims to identify and monitor the structure's modal parameters from the vibration response recorded by four accelerometers under unknown continuously varying excitation. Initially, the engineering aspects of SHM implementations are detailed and a state-of-the-art review about output-only modal identification techniques is presented, focusing on the Enhanced Frequency Domain Decomposition (EFDD) procedure. Then, a linear elastic Finite Element Model (FEM) of the structure is developed for the numerical extraction of modal parameters. Subsequently, the EFDD technique is implemented to identify modal frequencies and mode shapes from the recorded acceleration signals. The Continuous Wavelet Transform (CWT) time-frequency maps are analysed to distinguish between harmonic components and structural frequencies. Finally, the results of a first attempt SHM plan for the building are presented, referring to the data acquired over five subsequent months. The purpose is to monitor the evolution of modal parameters estimated through the EFDD procedure and to study the behaviour of the FEM in view of a future model updating.
Le infrastrutture civili rivestono una rilevante importanza sociale, economica e strategica. Molte di queste eccedono la loro vita utile di progetto e sono soggette a fenomeni di degrado ed eventi naturali imprevisti. In questo contesto, implementare efficaci strategie di monitoraggio strutturale è cruciale nel garantire adeguati livelli prestazionali, un'efficiente pianificazione degli interventi manutentivi e il prolungamento della vita utile delle strutture. Questa tesi si concentra sul monitoraggio di un edificio industriale esistente, soggetto all'eccitazione prodotta da macchine per lo stampaggio ad iniezione. I loro cicli di lavoro generano componenti armoniche a frequenze simili a quelle che caratterizzano il comportamento dinamico dell'edificio. L'analisi mira a identificare e monitorare i parametri modali strutturali attraverso la risposta vibrazionale registrata da quattro accelerometri sotto un'eccitazione variabile sconosciuta. Inizialmente, vengono discussi gli aspetti ingegneristici del monitoraggio strutturale e le tecniche di identificazione modale basate sulla sola misura della risposta strutturale, con focus sulla tecnica Enhanced Frequency Domain Decomposition (EFDD). Successivamente, viene sviluppato un modello elastico lineare ad elementi finiti per il calcolo numerico dei parametri modali. In seguito, la tecnica EFDD viene implementata per identificare frequenze e forme modali dai segnali di accelerazione registrati. Tramite la trasformata wavelet continua, le mappe tempo-frequenza vengono analizzate per distinguere tra componenti armoniche e frequenze strutturali. Infine, i risultati del primo piano di monitoraggio strutturale vengono presentati, in riferimento ai dati acquisiti in cinque mesi successivi. Lo scopo è monitorare l'evoluzione dei parametri modali stimati con la procedura EFDD e studiare il comportamento del modello in vista di una futura calibrazione.
Structural health monitoring of an industrial building through operational modal analysis techniques
Galeazzi, Tommaso
2022/2023
Abstract
Civil engineering structures hold relevant social, economic, and strategic importance. Many of them exceed their estimated design life and must withstand degradation, damage, and unforeseen natural events. In this context, implementing efficient Structural Health Monitoring (SHM) strategies ensures adequate performance levels, efficient maintenance interventions planning and extends structures' service life. This thesis focuses on the SHM of an existing industrial building subjected to the excitation produced by injection moulding machines. Their working cycles generate harmonic components at frequencies close to those characterising the building's dynamic behaviour. The analysis aims to identify and monitor the structure's modal parameters from the vibration response recorded by four accelerometers under unknown continuously varying excitation. Initially, the engineering aspects of SHM implementations are detailed and a state-of-the-art review about output-only modal identification techniques is presented, focusing on the Enhanced Frequency Domain Decomposition (EFDD) procedure. Then, a linear elastic Finite Element Model (FEM) of the structure is developed for the numerical extraction of modal parameters. Subsequently, the EFDD technique is implemented to identify modal frequencies and mode shapes from the recorded acceleration signals. The Continuous Wavelet Transform (CWT) time-frequency maps are analysed to distinguish between harmonic components and structural frequencies. Finally, the results of a first attempt SHM plan for the building are presented, referring to the data acquired over five subsequent months. The purpose is to monitor the evolution of modal parameters estimated through the EFDD procedure and to study the behaviour of the FEM in view of a future model updating.File | Dimensione | Formato | |
---|---|---|---|
2023_12_Galeazzi_Tesi_01.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
97.19 MB
Formato
Adobe PDF
|
97.19 MB | Adobe PDF | Visualizza/Apri |
2023_12_Galeazzi_Executive Summary_02.pdf
non accessibile
Descrizione: Executive summary
Dimensione
360.62 kB
Formato
Adobe PDF
|
360.62 kB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/214911